1. Field of the Invention
The present disclosure generally relates to the fabrication of integrated circuits, and, more particularly, to various methods of forming FinFET semiconductor devices using a replacement gate technique and the resulting semiconductor devices.
2. Description of the Related Art
In modern integrated circuits, such as microprocessors, storage devices and the like, a very large number of circuit elements, especially transistors, are provided and operated on a restricted chip area. Immense progress has been made over recent decades with respect to increased performance and reduced feature sizes of circuit elements, such as transistors. However, the ongoing demand for enhanced functionality of electronic devices forces semiconductor manufacturers to steadily reduce the dimensions of the circuit elements and to increase the operating speed of the circuit elements. The continuing scaling of feature sizes, however, involves great efforts in redesigning process techniques and developing new process strategies and tools so as to comply with new design rules. Generally, in complex circuitry including complex logic portions, MOS technology is presently a preferred manufacturing technique in view of device performance and/or power consumption and/or cost efficiency. In integrated circuits including logic portions fabricated by MOS technology, field effect transistors (FETs) are provided that are typically operated in a switched mode, that is, these devices exhibit a highly conductive state (on-state) and a high impedance state (off-state). The state of the field effect transistor is controlled by a gate electrode, which controls, upon application of an appropriate control voltage, the conductivity of a channel region formed between a drain region and a source region.
To improve the operating speed of FETs, and to increase the density of FETs on an integrated circuit device, device designers have greatly reduced the physical size of FETs over the years. More specifically, the channel length of FETs has been significantly decreased, which has resulted in improving the switching speed of FETs. However, decreasing the channel length of a FET also decreases the distance between the source region and the drain region. In some cases, this decrease in the separation between the source and the drain makes it difficult to efficiently inhibit the electrical potential of the source region and the channel from being adversely affected by the electrical potential of the drain. This is sometimes referred to as a so-called short channel effect, wherein the characteristic of the FET as an active switch is degraded.
In contrast to a FET, which has a planar structure, a so-called FinFET device has a three-dimensional (3D) structure. FIG. 1A is a perspective view of an illustrative prior art FinFET semiconductor device “A” that is formed above a semiconductor substrate B that will be referenced so as to explain, at a very high level, some basic features of a FinFET device. In this example, the FinFET device A includes three illustrative fins C, a gate structure D, sidewall spacers E and a gate cap layer F. The gate structure D is typically comprised of a layer of gate insulating material (not separately shown), e.g., a layer of high-k insulating material or silicon dioxide, and one or more conductive material layers (e.g., metal and/or polysilicon) that serve as the gate electrode for the device A. The fins C have a three-dimensional configuration: a height H, a width W and an axial length L. The axial length L corresponds to the direction of current travel in the device A when it is operational. The portions of the fins C covered by the gate structure D are the channel regions of the FinFET device A. In a conventional process flow, the portions of the fins C that are positioned outside of the spacers E, i.e., in the source/drain regions of the device A, may be increased in size or even merged together (a situation not shown in FIG. 1A) by performing one or more epitaxial growth processes. The process of increasing the size of or merging the fins C in the source/drain regions of the device A is performed to reduce the resistance of source/drain regions and/or make it easier to establish electrical contact to the source drain regions. Even if an epi “merger” process is not performed, an epi growth process will typically be performed on the fins C to increase their physical size.
In the FinFET device A, the gate structure D may enclose both sides and the upper surface of all or a portion of the fins C to form a tri-gate structure so as to use a channel having a three-dimensional structure instead of a planar structure. In some cases, an insulating cap layer (not shown), e.g., silicon nitride, is positioned at the top of the fins C and the FinFET device only has a dual-gate structure (sidewalls only). Unlike a planar FET, in a FinFET device, a channel is formed perpendicular to a surface of the semiconducting substrate so as to reduce the physical size of the semiconductor device. Also, in a FinFET, the junction capacitance at the drain region of the device is greatly reduced, which tends to significantly reduce short channel effects. When an appropriate voltage is applied to the gate electrode of a FinFET device, the surfaces (and the inner portion near the surface) of the fins C, i.e., the vertically oriented sidewalls and the top upper surface of the fin, form a surface inversion layer or a volume inversion layer that contributes to current conduction. In a single fin FinFET device, the “channel-width” is estimated to be about two times (2×) the vertical fin-height plus the width of the top surface of the fin, i.e., the fin width. Multiple fins can be formed in the same foot-print as that of a planar transistor device. Accordingly, for a given plot space (or foot-print), FinFETs tend to be able to generate significantly higher drive current density than planar transistor devices. Additionally, the leakage current of FinFET devices after the device is turned “OFF” is significantly reduced as compared to the leakage current of planar FETs, due to the superior gate electrostatic control of the “fin” channel on FinFET devices. In short, the 3D structure of a FinFET device is a superior MOSFET structure as compared to that of a planar FET, especially in the 20 nm CMOS technology node and beyond. The gate structures D for such FinFET devices may be manufactured using so-called “gate-first” or “replacement gate” (gate-last) manufacturing techniques.
For many early device technology generations, the gate structures of most transistor elements (planar or FinFET devices) were comprised of a plurality of silicon-based materials, such as a silicon dioxide and/or silicon oxynitride gate insulation layer, in combination with a polysilicon gate electrode. However, as the channel length of aggressively scaled transistor elements has become increasingly smaller, many newer generation devices employ gate structures that contain alternative materials in an effort to avoid the short channel effects which may be associated with the use of traditional silicon-based materials in reduced channel length transistors. For example, in some aggressively scaled transistor elements, which may have channel lengths on the order of approximately 10-32 nm or less, gate structures that include a so-called high-k dielectric gate insulation layer and one or more metal layers that function as the gate electrode (HK/MG) have been implemented. Such alternative gate structures have been shown to provide significantly enhanced operational characteristics over the heretofore more traditional silicon dioxide/polysilicon gate structure configurations.
Depending on the specific overall device requirements, several different high-k materials—i.e., materials having a dielectric constant, or k-value, of approximately 10 or greater—have been used with varying degrees of success for the gate insulation layer in an HK/MG gate electrode structure. For example, in some transistor element designs, a high-k gate insulation layer may include tantalum oxide (Ta2O5), hafnium oxide (HfO2), zirconium oxide (ZrO2), titanium oxide (TiO2), aluminum oxide (Al2O3), hafnium silicates (HfSiOx) and the like. Furthermore, one or more non-polysilicon metal gate electrode materials—i.e., a metal gate stack—may be used in HK/MG configurations so as to control the work function of the transistor. These metal gate electrode materials may include, for example, one or more layers of titanium (Ti), titanium nitride (TiN), titanium-aluminum (TiAl), titanium-aluminum-carbon (TiALC), aluminum (Al), aluminum nitride (AlN), tantalum (Ta), tantalum nitride (TaN), tantalum carbide (TaC), tantalum carbonitride (TaCN), tantalum silicon nitride (TaSiN), tantalum silicide (TaSi) and the like.
One well-known processing method that has been used for forming a transistor with a high-k/metal gate structure is the so-called “gate last” or “replacement gate” technique. The replacement gate process may be used when forming planar devices or 3D devices. FIGS. 1B-1E simplistically depict one illustrative prior art method for forming an HK/MG replacement gate structure using a replacement gate technique on a planar transistor device. As shown in FIG. 1B, the process includes the formation of a basic transistor structure above a semiconducting substrate 12 in an active area defined by a shallow trench isolation structure 13. At the point of fabrication depicted in FIG. 1A, the device 10 includes a sacrificial gate insulation layer 14, a dummy or sacrificial gate electrode 15, sidewall spacers 16, a layer of insulating material 17 and source/drain regions 18 formed in the substrate 12. The various components and structures of the device 10 may be formed using a variety of different materials and by performing a variety of known techniques. For example, the sacrificial gate insulation layer 14 may be comprised of silicon dioxide, the sacrificial gate electrode 15 may be comprised of polysilicon, the sidewall spacers 16 may be comprised of silicon nitride and the layer of insulating material 17 may be comprised of silicon dioxide. The source/drain regions 18 may be comprised of implanted dopant materials (N-type dopants for NMOS devices and P-type dopants for PMOS devices) that are implanted into the substrate 12 using known masking and ion implantation techniques. Of course, those skilled in the art will recognize that there are other features of the transistor 10 that are not depicted in the drawings for purposes of clarity. For example, so-called halo implant regions are not depicted in the drawings, as well as various layers or regions of silicon/germanium that are typically found in high performance PMOS transistors. At the point of fabrication depicted in FIG. 1B, the various structures of the device 10 have been formed and a chemical mechanical polishing (CMP) process has been performed to remove any materials above the sacrificial gate electrode 15 (such as a protective cap layer (not shown) comprised of silicon nitride) so that at least the sacrificial gate electrode 15 may be removed.
As shown in FIG. 1C, one or more etching processes are performed to remove the sacrificial gate electrode 15 and the sacrificial gate insulation layer 14 to thereby define a gate cavity 20 where a replacement gate structure will subsequently be formed. Typically, the sacrificial gate insulation layer 14 is removed as part of the replacement gate technique, as depicted herein. However, the sacrificial gate insulation layer 14 may not be removed in all applications.
Next, as shown in FIG. 1D, various layers of material that will constitute a replacement gate structure 30 are formed in the gate cavity 20. Even in cases where the sacrificial gate insulation layer 14 is intentionally removed, there will typically be a very thin native oxide layer (not shown) that forms on the substrate 12 within the gate cavity 20. The materials used for the replacement gate structures 30 for NMOS and PMOS devices are typically different. For example, the replacement gate structure 30 for an NMOS device may be comprised of a high-k gate insulation layer 30A, such as hafnium oxide, having a thickness of approximately 2 nm, a first metal layer 30B (e.g., a layer of titanium nitride with a thickness of about 1-2 nm), a second metal layer 30C—a so-called work function adjusting metal layer for the NMOS device—(e.g., a layer of titanium-aluminum or titanium-aluminum-carbon with a thickness of about 5 nm), a third metal layer 30D (e.g., a layer of titanium nitride with a thickness of about 1-2 nm) and a bulk metal layer 30E, such as aluminum or tungsten.
Ultimately, as shown in FIG. 1E, one or more CMP processes are performed to remove excess portions of the gate insulation layer 30A, the first metal layer 30B, the second metal layer 30C, the third metal layer 30D and the bulk metal layer 30E positioned outside of the gate cavity 20 to thereby define the replacement gate structure 30 for an illustrative NMOS device. Typically, the replacement gate structure 30 for a PMOS device does not include as many metal layers as does an NMOS device. For example, the gate structure 30 for a PMOS device may only include the high-k gate insulation layer 30A, a single layer of titanium nitride—the work function adjusting metal for the PMOS device—having a thickness of about 3-4 nm, and the bulk metal layer 30E.
FIG. 1F depicts the device 10 after several process operations were performed. First, one or more etching processes were performed to remove upper portions of the various materials within the cavity 20 so as to form a recess within the gate cavity 20. Then, a gate cap layer 31 was formed in the recess above the recessed gate materials. The gate cap layer 31 is typically comprised of silicon nitride and it may be formed by depositing a layer of gate cap material so as to over-fill the recess formed in the gate cavity and, thereafter, performing a CMP process to remove excess portions of the gate cap material layer positioned above the surface of the layer of insulating material 17. The gate cap layer 31 is formed so as to protect the underlying gate materials during subsequent processing operations.
As the gate length of transistor devices has decreased, the physical size of the gate cavity 20 has also decreased. Thus, it is becoming physically difficult to fit all of the layers of material needed for the replacement gate structure 30, particularly for NMOS devices, due to the greater number of layers of material that are typically used to form the gate structures for the NMOS devices, within the reduced-size gate cavity. For example, as gate lengths continue to decrease, voids or seams may be formed as the various layers of material are deposited into the gate cavity 20. FIG. 1G is a somewhat enlarged view of an illustrative NMOS device that is provided in an attempt to provide the reader with some idea of just how limited the lateral space 20S is within the gate cavity 20 of an NMOS device as the various metal layers 30A-30D are formed in the gate cavity 20. In FIG. 1G, the internal sidewall surfaces of the spacers 16 define a gate cavity 20 having a substantially uniform width 20S throughout the height or depth of the gate cavity 20. As the layers of material in the gate stack for the device are formed in the cavity 20, the remaining space 39 within the gate cavity 20 becomes very small. As the latter metal layers are formed, the lateral space 39 may be about 1-2 nm in width or even smaller. In some cases, the space 39 may be essentially non-existent. This may lead to so-called “pinch-off” of metal layers such that voids or seams may be formed in the overall gate stack, which may result in devices that perform at levels less than anticipated or, in some cases, the formation of devices that are simply not acceptable and have to be discarded. Additionally, an etch-back process is traditionally performed on the layers 30A-D to make room within the upper portion of the gate cavity 20 for a bulk conductor material, such as tungsten and aluminum, and a gate cap layer. During this recess etching process, some form of a protective material must be formed in the gate cavity 20 above the metal layer 30D to protect desired portions of the underlying metal layers during the recess etching process. If the lateral space 39 (to the extent it exists) cannot be reliably filled with such a protective material, such as a flowable oxide material, then the recessing etching process cannot be performed for fear of removing undesired portions of the metal layers during the course of performing the recess etching process.
Over recent years, due to the reduced dimensions of the transistor devices, the operating speed of the circuit components has been increased with every new device generation, and the “packing density,” i.e., the number of transistor devices per unit area, in such products has also increased during that time. Such improvements in the performance of transistor devices has reached the point where one limiting factor relating to the operating speed of the final integrated circuit product is no longer the individual transistor element but the electrical performance of the complex wiring system that is formed above the device level that includes the actual semiconductor-based circuit elements. Typically, due to the large number of circuit elements and the required complex layout of modern integrated circuits, the electrical connections of the individual circuit elements cannot be established within the same device level on which the circuit elements are manufactured, but require one or more additional metallization layers, which generally include metal-containing lines providing the intra-level electrical connection, and also include a plurality of inter-level connections or vertical connections, which are also referred to as vias. These vertical interconnect structures comprise an appropriate metal and provide the electrical connection of the various stacked metallization layers.
Furthermore, in order to actually connect the circuit elements formed in the semiconductor material with the metallization layers, an appropriate vertical contact structure is provided, a first end of which is connected to a respective contact region of a circuit element, such as a gate electrode and/or the drain and source regions of transistors, and a second end that is connected to a respective metal line in the metallization layer by a conductive via. As device dimensions have decreased, the conductive contact elements in the contact level have to be provided with critical dimensions in the same order of magnitude. The contact elements typically represent plugs, which are formed of an appropriate metal or metal composition, wherein, in sophisticated semiconductor devices, tungsten, in combination with appropriate barrier materials, has proven to be a viable contact metal. For this reason, contact technologies have been developed in which contact openings are formed in a self-aligned manner by removing dielectric material, such as silicon dioxide, selectively from the spaces between closely spaced gate electrode structures. That is, after completing the transistor structure, the gate electrode structures are used as etch masks for selectively removing the silicon dioxide material in order to expose the source/drain regions of the transistors, thereby providing self-aligned trenches which are substantially laterally delineated by the spacer structures of the gate electrode structures.
However, the aforementioned process of forming self-aligned contacts results in an undesirable loss of the materials that protect the conductive gate electrode, i.e., the gate cap layer and the sidewall spacers, as will be explained with reference to FIGS. 2A-2B. FIG. 2A schematically illustrates a cross-sectional view of an integrated circuit product 40 at an advanced manufacturing stage. As illustrated, the product 40 comprises a plurality of illustrative gate structures 41 that are formed above a substrate 42, such as a silicon substrate. The gate structures 41 are comprised of an illustrative gate insulation layer 43 and an illustrative gate electrode 44 that are formed in a gate cavity 20 using a gate-last processing technique. An illustrative gate cap layer 46 and sidewall spacers 48 encapsulate and protect the gate structures 41. The gate cap layer 46 and sidewall spacers 48 are typically made of silicon nitride. Also depicted in FIG. 2A are a plurality of raised source/drain regions 50 and a layer of insulating material 52, e.g., silicon dioxide. FIG. 2B depicts the product 40 after a contact opening 54 has been formed in the layer of insulating material 52 for a self-aligned contact. Although the contact etch process performed to form the opening 54 is primarily directed at removing the desired portions of the layer of insulating material 52, portions of the protective gate cap layer 46 and the protective sidewall spacers 48 get consumed during the contact etch process, as simplistically depicted in the dashed regions 56. Given that the cap layer 46 and the spacers 48 are attacked in the contact etch process, the thickness of these protective materials must be sufficient such that, even after the contact etch process is completed, there remains sufficient cap layer material and spacer material to protect the gate structures 41. Accordingly, device manufacturers tend to make the cap layers 46 and spacers 48 “extra thick,” i.e., with an additional thickness that may otherwise not be required but for the consumption of the cap layers 46 and the spacers 48 during the contact etch process. In turn, increasing the thickness of such structures, i.e., increasing the thickness of the gate cap layers 46, causes other problems, such as increasing the aspect ratio of the contact opening 54 due to the increased height, increasing the initial gate height, which makes the gate etching and spacer etching processes more difficult, etc.
As device dimensions continue to shrink, self-aligned contacts are needed to prevent an electrical short between the gate and the source/drain contact element. As noted above, in prior art processing techniques, a relatively thick gate cap layer 46 (e.g., silicon nitride) is formed to protect the underlying metal layers in the gate electrode 44 from being exposed during the source/drain contact etch process which results in the formation of the contact opening 54 (see FIGS. 2A-2B). If the gate electrode materials are exposed during the contact etch process, then an electrical short will be formed between the gate electrode 44 and the source/drain contact element (not shown) that will be formed in the contact opening 54. There are two ways to “make room” in the gate cavity 20 for the gate cap layer 46 by recessing the work function metal materials within the gate cavity 20. A first approach involves filling the remaining portions of the gate cavity 20 with a relatively thick work function metal until it (collectively) is “pinched-off” within the gate cavity 20, and thereafter performing an etching process to recess it down, thereby making room for the gate cap layer 46. This approach is very challenging due to the fact that the work function metals are typically comprised of multiple layers of metal (e.g., TiN/TiC/TiN. TiN/TaC/TiN, etc.) that are layered within the gate cavity 20 so that a uniform recess is not easy to form by a single etch-back process. Also, due to the “pinch-off” that typically occurs when forming such layers of metal, there are inevitable void/seam formations due to the very small size of the gate cavity as these material layers are formed, which further increases the difficulty of the etch-back process that is performed to recess the various metal layers. A second approach involves formation of relatively thinner work function metal layer(s) so as to not “pinch-off” the gate cavity 20, then fill in the sacrificial material, such as OPL or a flowable oxide, that can reliably fill the remaining portions of the gate cavity 20. Then, the sacrificial material may be recessed so as to expose the desired portions of the work function metal to be removed, the exposed portions of the work function metals are then removed by etching while the remaining portions of the sacrificial material protect the underlying metal layers. Thereafter, the sacrificial material may be removed from within the gate cavity 20. While this latter approach tends to produce better results, as gate lengths continue to shrink (e.g., 15 nm or less), the required work function metal thickness (e.g., 7 nm) simply fills too much of the gate cavity, with the result being that there is little, if any, space left within the gate cavity 20 for the sacrificial material that is used in the recess etching process.
The present disclosure is directed to various methods of forming FinFET semiconductor devices using a replacement gate technique and the resulting semiconductor devices that may avoid, or at least reduce, the effects of one or more of the problems identified above.